magic-linguistic-eval

Installation
SKILL.md

When to Use

  • Reporting quality numbers for any non-English LLM.
  • Choosing benchmark + metric for a language pair / task.
  • Building grammatical-knowledge probes (BLiMP-style).
  • Contamination audit (cross-reference magic-linguistic-corpus).
  • Adding fairness eval (per-dialect / per-register breakdown).

When NOT to use: purely qualitative review → human review pipeline. Per-dataset license check → magic-linguistic-ethics. Tokenizer fertility audit → magic-linguistic-tokenize.

Why this is A-tier

The Evaluate phase is the orchestrator's last specialist. Eval results drive release decisions. Picking BLEU over chrF for a morphologically-rich language doesn't just produce one bad number — it produces a misleading number that drives wrong investment decisions for months. Eval is leverage; weak eval cascades.

The Knowledge Engineers Routinely Miss

  1. BLEU is pathological for morphologically-rich languages. Single-morpheme edits wreck it as harshly as full mistranslations. For Turkish / Finnish / Swahili / Yoruba / Inuktitut: chrF / chrF++ / spBLEU are primary. Report BLEU as supplementary if at all — never as the headline.

  2. COMET coverage varies wildly. COMET-22 has good European + Indic coverage; spotty Bantu / Indigenous Americas. xCOMET-XL extends but still has gaps. ALWAYS check per-language coverage before reporting; a "missing" language gets random numbers.

Installs
2
GitHub Stars
1
First Seen
Jul 3, 2026
magic-linguistic-eval — votee-ai/magic-agent-skills